Professor Graham Hatfull and others in lab together
Features & Articles

Bacteria-killing viruses discovered by Pitt researchers are saving patients who have no other options

  • Technology & Science
  • Innovation and Research
  • Kenneth P. Dietrich School of Arts and Sciences

Antibiotics revolutionized medicine, but even these lifesaving medications have their limits. In two new studies involving University of Pittsburgh biologists, researchers show that an experimental technique using bacteria-killing viruses successfully treated two patients with deadly infections that were unresponsive to antibiotics.  

“One case study is anecdotal. It’s a one-off: We might have gotten lucky,” said Graham Hatfull, the Eberly Family Professor of Biotechnology in the Kenneth P. Dietrich School of Arts and Sciences. With three documented successful cases, “now we can start to look for consistencies and some sense that it may work for many more patients.”

Bacteriophages, phages for short, kill bacteria by injecting their DNA into bacterial cells and turning their targets into phage factories. Harmless to humans, that ability has made phages a candidate for treating some otherwise intractable infections — including those caused by Mycobacterium abscessus, a species of bacteria that’s notoriously resistant to antibiotics and can be deadly to patients with cystic fibrosis or who are on drugs that suppress their immune systems.

Hatfull and his lab showed the technique’s potential in 2019 in the first use of phages to treat a Mycobacterium infection and the first use of genetically engineered phages in such a treatment. They’re now on a path toward helping more patients who have reached the limits of what established medicine can provide.

“You have to have exhausted all of those options, from a regulatory point of view, before you can go to something truly experimental like phages,” said Hatfull. “These people don’t have anybody else to go to. It’s us or nothing.”

These people don’t have anybody else to go to. It’s us or nothing.

Graham Hatfull

A lifesaving transplant

In a paper published today in Cell, researchers report that the Hatfull Lab’s phage therapy was successful on another, much more common infection in patients with cystic fibrosis — a disease that causes buildups of mucus in the lungs, among other catastrophic effects.

The success of the study shows that phage therapy may be useful for many more cystic fibrosis patients, potentially helping others who are stuck on transplant waitlists due to stubborn mycobacterial infections in the lungs.

But the treatment’s success was anything but certain. Performing phage therapy on a patient with a lung infection, rather than a skin infection like in the 2019 paper, brought with it a new set of challenges and questions.

“If there’s bacteria in the bloodstream, and you give a phage IV, you’ve got a pretty good chance that those two will find each other,” Hatfull said. The lungs, however, are a complicated network of tubes where an infection can hide. “It’s just a big unknown, essentially.”

That’s especially true in patients like Jarrod Johnson, at the center of the new paper. Johnson’s infection persisted even after years of treatment with antibiotics. After phage therapy, however, he tested negative for the infection more than a dozen times, clearing the way for him to receive a lung transplant.

“I am so grateful for the effort, persistence and creativity of all the people who were involved in my treatment,” said Johnson. “I thought I was going to die. They have literally saved my life.”  

The work was carried out by physicians at National Jewish Health in Denver, Colorado, and led by Jerry Nick, who sent samples to Hatfull’s lab to seek phages that could tackle the infection.

“We’ve got a large collection of phages, and we’ve sequenced over 4,000 of their genomes, so we understand their genomic profiles and relationships in exquisite detail,” Hatfull said. “Plus, we have information about which bacteria are infected by which phages, which means that we can take intelligent guesses as to which are going to be the most useful types of phages to look for.”

When they received the request, Hatfull and his lab were able to identify two phages that matched up with the strain of Mycobacterium abscessus in Johnson’s lungs and genetically engineer them to be more effective at killing it.

As with any case study, it’s impossible to be sure that the patient’s recovery was a result of the treatment. But the evidence in this case is stronger than most, Hatfull said.

“The patient has been followed incredibly carefully, clinically, for years, with multiple sequencing of strains and antibiotic profiles,” he said. Especially convincing were molecular signals of parts of the bacteria’s cell wall in the patient’s urine a sign that the phages were killing the bacteria and doing their job.

From a cocktail to a straight shot

In the second of the two studies, published last week in Nature Communications, the Pitt team, along with colleagues at Harvard University led by Jessica Little, showed success in treating an immunocompromised patient with a mycobacterial skin infection.

“The therapy seemed to really make a big difference when the antibiotics really hadn’t done much for a long period of time,” said Hatfull.

It’s the first instance of phage therapy used to treat an infection by Mycobacterium chelonae — a related, but different species from the bacteria previously treated with phages. And while the patient’s immune system seemed to mount an attack against the phage, it didn’t stop the treatment from being effective.

Yet another novel aspect of the study was the phage the researchers used.

Researchers worry about the potential for bacteria to evolve resistance to viruses. Delivering a cocktail of multiple phages is how Hatfull’s team previously attempted to avoid the problem: Barrage a bacterial infection with three kinds of viruses at the same time, and it’s less likely to be able to out-evolve them.

But when the team tested phages against this patient’s infection, only one turned out to be effective: Muddy, one of the three used in the 2019 study.

“That’s all that we had — we used all of the phages that were available,” Hatfull said. “We obviously ran the risk that we would see resistance and failure. But it was either that or do nothing.”

The fact that the bacteria didn’t evolve resistance to the single phage came as a surprise, and it also opens up new potential for treating patients, especially for those with a robust immune system.

“Now we can contemplate giving them phage number one for a month. And then we’ll stop. Then give them phage number two for a month and then stop,” Hatfull said. Doing so would lessen the risk of the patient’s immune system attacking the phages.

Taken together, the two studies point toward a future for phage therapy for a broader array of Mycobacterium infections and new potential ways to maximize the therapy’s chance of success.

Opening the floodgates

Until today, it may have seemed to an outside observer that the lab’s clinical work started and ended with that 2019 breakthrough. The truth is just the opposite.

“That’s when the floodgates opened,” said Rebekah Dedrick, a research associate in Hatfull’s lab. “We started to get requests from around the world, and we still get them.”

Dedrick coordinates the lab’s clinical side, handling the mountain of permits and other paperwork required to give patients around the world an experimental therapy. She and a team of three research technicians also screen samples of bacteria sent from those patients, test viruses against them to see if any may be effective and perform follow-up testing after the treatment.

It was a sudden shift for Dedrick, who up to that point in her career had focused on basic science. But she quickly embraced it. “There’s nothing else for the clinicians to give them, and then we offer this little nugget of hope,” she said. “Just being able to offer that to a patient who has no other options, it’s really fulfilling.”

Other members of the lab have also become an important part of the team’s clinical work, including Research Instructor Deborah Jacobs-Sera, who cultures and purifies the phages to a quantity and quality that they can be injected into patients. What was first a nerve-wracking process has through years of practice become a well-oiled machine that takes in samples from patients and sends out lifesaving therapies.

Viruses hiding in DNA

Experimental medicine, however, is just one component of the lab’s work. At the same time, the group is continuing basic research into what makes these viruses and bacteria tick, producing clues about ways to improve future treatments.

One target: the samples of Mycobacterium cells sent by more than 270 patients looking for treatment.

“As well as having a really good and broad and diverse collection of phages, we now have a big, broad and diverse collection of clinical isolates of Mycobacterium abscessus and some related strains,” Hatfull said. “This is a huge resource for understanding basic biology.”

In that library of samples, Hatfull and his lab have found a surprising new place to look for candidate phages to enlist in the fight against Mycobacterium infections: Hidden in the DNA of these bacteria are the full genetic material of phages that infected those cells in the past. Their existence is proof that the phage can infect those bacteria, and the lab is now able to use these blueprints to bring these viruses to life. The next step is to tweak their DNA so they can not only infect those Mycobacterium strains, but reliably kill them, too.

“We’re still working on how to engineer them and how to get them into a therapy,” said Hatfull. “We haven’t used any of them therapeutically yet, but I think we’re getting close.”

And continuing research on which phages infect which Mycobacterium strains and why could impact treatments in the future. At the heart of that question is the potential to move beyond a time- and labor-intensive personalized medicine approach to a more generalized treatment that could help patients on a much broader scale.

The three-legged stool

The lab’s educational mission has forged ahead too, even during a pandemic. Since 2002, Hatfull has led a program that involves undergraduates in cutting-edge research. Now, this program is called the Howard Hughes Medical Institute Science Education Alliance-Phage Hunters Advancing Genomic and Evolutionary Science (HHMI SEA-PHAGES), and it gives students the chance to discover and name their own viruses. One such virus discovered through the program was Muddy, now used in around half of the lab’s therapies.

Hatfull refers to the three aims of his lab as a “three-legged stool,” with each leg — research, education and clinical work — enriching the others.

“We have this constant, ongoing process of expanding the number of phages that we have available to us,” he said. At the same time, 30 years of work on the genetics of phages allows them to effectively drink from the firehose of phages, organizing them in such a way that they’re ready for quick assessments when a mycobacterium sample arrives.

As for their clinical work, the lab expects to soon publish reports of more case studies, moving past proof of concept and developing a more robust body of knowledge about phage therapy for Mycobacterium infections — the kind that’s essential to lay the groundwork for a potential clinical trial where the therapy could be tested in a more rigorous way.

“We need clinical trials, and we aren't going to get there until we have some information about each of these steps,” said Dedrick.

Wherever the field goes next, it’s all but certain that the lab will be at the center of it.

“When it comes to people who work on phages that could be used for mycobacterial diseases, we are it, basically,” said Hatfull. “You can’t get this from any other lab in the world, I think it’s fair to say.”


Patrick Monahan, photos by Aimee Obidzinski